At the time of writing this document the entire planet is facing the pandemic caused by the SARS-CoV-2 virus. Contrary to the initial thoughts that most infected patients had symptoms, according to data from China collected in April 2020, at the time of the diagnosis, up to 80% of those confirmed to have the disease are asymptomatic, becoming an important source of contagion.1–3 Pan et al. reported that health workers in Wuhan (China) had a significantly higher risk of becoming infected (daily confirmed case rate in local health care workers was 130.5 per million while in the general population it was 41.5 per million).4

In this light, the scientific community is discussing the use of chemoprophylaxis in people at higher risk of infection using several alternatives including antimalarials (chloroquine or hydroxychloroquine) and antiretrovirals (lopinavir-ritonavir).5–10 There is an increased interest in use of chloroquine and hydroxychloroquine, two medications that have experimentally shown to have antiviral capabilities and prophylactic potential.11,12 Lee et al. recently reported good results in an observational study in South Korea after a large COVID-19 exposure event in a hospital. At the end of quarantine, after receiving post-exposure prophylaxis with hydroxychloroquine (400mg daily for 14 days) all follow-up PCR tests were negative in 211 individuals exposed to the index case.13 Certain countries have already adopted chemoprophylaxis schemes; on March 22nd, 2020, the Indian Council of Medical Research's National Task Force for COVID-19 issued a national recommendation to use hydroxychloroquine for prophylaxis against SARS-CoV-2 infection (400mg twice a day on day 1, followed by 400mg once weekly for 7 weeks).14

We conducted a search (updated to April 15th, 2020) on the website https://clinicaltrials.gov/. Using the keywords COVID-19 and hydroxychloroquine we found a total of 90 projects registered. Twenty-five of those projects included the use of prophylaxis in non-infected population (Table 1)

Institutions from 13 countries are leading those projects, 13 of them from the United States, 2 from Mexico, 2 from Spain and 2 from France. Turkey, Colombia, Austria, South Korea, Singapore, United Kingdom, Thailand, Australia and Canada, have institutions leading one project (there are more institutions than projects, because some have shared leadership).

There is a significant variability in the number of participants among the studies. The estimated number of participants to be enrolled ranges between 45 and 55,000 with an average of 5588±13,139.2, and a median of 1212 participants. Only 3 (12%) studies plan to enroll more than 4000 participants; those 3 studies will include 70,000 participants which corresponds to 50.1% of the total potential recruitment of the 25 protocols (NCT04334148, NCT04303507 and NCT04333732).

There is a significant variability among protocols regarding hydroxychloroquine maintenance dose, which will be between 200 and 600mg. The frequency is also highly variable: seventeen protocols will use daily prophylaxis for a period from 4 days to 12 weeks and 9 protocols plan to evaluate weekly use for a period of 3 to 24 weeks. Thirteen (52% of 25) protocols will use an initial loading dose ranging between 400–1400mg taken on the first day. In three other protocols, 2–4 days of loading doses of 400mg/day will be indicated.

19 clinical trials will evaluate pre-exposure prophylaxis and 6 post-exposure prophylaxis. In 9 of the pre-exposure prophylaxis studies and 4 of the post-exposure prophylaxis studies, a loading dose of 800mg of hydroxychloroquine will be started on the first day. In an additional multicenter pre-exposure prophylaxis study, which plans to recruit 15,000 participants (NCT04334148), they will use a higher loading dose on the first day: 1200mg of hydroxychloroquine.

We evaluated the exclusion criteria among protocols by grouping into several possible categories. Most common criteria used by protocols to exclude patients comprised allergies to 4-aminoquinolines (hydroxychloroquine, chloroquine) in 20 studies (80%); retinopathy in 19 (76%); history of a prolonged QT syndrome, use of medications that prolong the QT/QTc interval or risk factors for torsades de pointe in 17 (68%); nephropathy in 16 (64%); pregnancy or breastfeeding in 14 (56%); concomitant use of other medications with potential pharmacological interaction with 4-aminoquinolines in 13 (52%); liver disease in 13 (52%); psoriasis or porphyria in 11 (44%) and glucose-6-phosphate dehydrogenase deficiency in 9 protocols (36%).

We used only clinicaltrials.gov to search the trials, which is a weakness of our study, because several protocols may be lost (those registered on https://www.clinicaltrialsregister.eu/, http://www.chictr.org.cn/index.aspx or https://apps.who.int/trialsearch/). However, due to the rapidity of the events related to the COVID-19 pandemic, we consider that the initial publication of the review including only those registered on clinicaltrials.gov is warranted, in order to give clinicians a preliminary picture.

According to the World Health Organization, as of April 15th, 2020, there are almost 2 million confirmed cases of SARS-CoV-2 virus in 213 countries and territories, but this is a underestimation, because as mentioned, around 80% of the infected people could be asymptomatic and go undetected.4 A high infection rate among health care workers would not only exacerbate the impending shortage of health care facilities and health personnel but would also increase the possibility of a more widespread dissemination.5 However, the current scientific evidence is still not conclusive for institutions and governments to adopt a general recommendation regarding the prophylactic use of hydroxychloroquine chemoprophylaxis in healthcare workers.

Chloroquine and its derivatives (e.g., hydroxychloroquine) have been used for malaria and autoimmune rheumatic diseases for almost 80 years, and both the data from the literature and the experience of clinicians show a low incidence of side effects, which are generally mild to moderate. Retinal toxicity, a serious effect, is related to long-term cumulative dose, and is rarely seen in short-term use (i.e. several weeks).9,15

These arguments perhaps tip the balance in favor of using prophylaxis for SARS-CoV-2 virus infection, as long as they are not contraindicated. Nonetheless we are yet to know the results of these clinical trials.